20091022 arb minutes

Architecture Review Board

October 22, 2009

Back to Agenda-minutes

Attendance

Name	Present	With	Affiliation	E-mail address
Aneja, Paul	Yes	Guest
Bond, Andy	?	ArB
Curry, Jane	Yes	ArB	Health Information Strategies	janecurry@healthinfostrategies.com
Grieve, Grahame	Yes	ArB	Kestral Computing	grahame@kestral.com.au
Hufnagle, Steve	?	Guest
Julian, Tony	Yes	ArB	Mayo Clinic	ajulian@mayo.edu
Koisch, John	?	ArB	NCI	koisch_john@bah.com
Loyd, Patrick	?	ARB	Gordon point Informatics LTD.	patrick.loyd@gpinformatics.com
Lynch, Cecil	Yes	ArB	ontoreason LLC	clynch@ontoreason.com
Mead, Charlie	Yes	ArB	Booz Allen Hamilton	charlie.mead@booz.com
Nelson, Dale	?	Arb	II4SM	dale@zed-logic.com
Ocasio, Wendell	?	ArB	Agilex Technologies	wendell.ocasio@agilex.com
Parker, Ron	Yes	ArB	CA Infoway	rparker@eastlink.ca
Quinn, John	?	ArB	Health Level Seven, Inc.	jquinn@HL7.org
Robertson, Scott	?	Guest	Kaiser Permanente
Shakir, Abdul-Malik	?	ArB	Shakir Consulting	ShakirConsulting@cs.com
Smith, Karen	Yes	Guest	Technical Editor	karen@smithspot.com
Thompson, Cliff	Yes	Guest	OntoSolutions LLC	cliff@ontosolutions.com
Ocasio, Wendell	Yes	Guest

Agenda

• Call to order
• Approval of Agenda
• Approval of minutes of the October 15, 2009 telcon.
  • approved by affirmation
• Review of DITA topics
• Vocab email re: balloting the MIF
• Clarification on the new project scope statement for HSSP process in agreement with OMG and impact to the ArB.
• Other business and planning for next meeting
• Adjournment

Call to order

The meeting was called to order at 6:00pm U.S. EDT by Chair Ron Parker with Tony Julian as Scribe. Quorum was acheived.

Approval of Agenda

The agenda was approved by affirmation.

Approval of minutes of the October 15, 2009October 15, 2009 telcon.

The minutes of the October 15, 2009 telcon were approved by affirmation

Review of DITA topics

Ron G Parker: Karen, we are ready for your topic...

Ron G Parker: You are presenting

Ron G Parker: Yes

Wendell Ocasio Joined

Karen Smith: now can you see my screen? OK, welcome everyone. I'm presenting the high-level outline of the SAEAF DITA book. It still follows the order that the ArB proposed.

Karen Smith: My goal is to get agreement on the high level organization at this meeting. Then I would like you guys to review the topic maps and send me your comments. Within a week or two I will be sending out detailed outlines for review

Ron Parker: Let’s take a look at the commentary:

Paul Aneja joined

Karen Smith: Regarding the BF and IF sections ---

Karen Smith: John K suggested putting the BF and IF sections after the Governance and ECCF sections. Do you agree with that suggestion?

Grahame Grieve: Grahame agrees with Jane

Jane Curry: Seems the order should be ECCF, governance, BF

Karen Smith: Yes that makes sense to me

Jane Curry: Because the information is the most familiar to people, and wont be done til the next phase, we should put it to last. Can we do it that way?

Ron Parker: Reasonable to me. Any Comments?

John K: Is this what is suggested?

Karen Smith: y\Yes, the new sections (in red) should come last. For the first phase I would have just placeholders for the IF, IG, and Examples section. Those will be filled in for phase 2. Okay?

Jane Curry: There is not sequence in it inherent to DITA, what should it be?

Jane Curry: There can be more than one roadmap into DITA.

Ron G Parker: Yes, Karen, the placeholders you suggest are fine.

Karen Smith: true

Paul Aneja: Would there be one document in the end, or separate?

Jane Curry: Now there is, but the goal is to make all of it in one place, and that you can produce one or 5 documents on how you want to look at it.

Jane Curry: Goal on phase 1 is to pull it together and make it coherent.

Karen Smith: For phase 1, I would have one document (the whole enchilada), but in phase 2, you can have multiple views (or documents) for different audiences -- architects, executives, HL7 chairs. Even in phase 1, you would be able to publish one of themajor subsections such as ECCF by itself if you want

Paul Aneja: Add links to othere sections, is this easy to do?

Karen Smith: Yes, phase 1 is to pull it together, and later phases will focus more on views, refining things, adding new stuff. Yes, adding links is usually easy. You can navigate the map sequentially like in a book or jump from topic to topic in a network.

Jane Curry: A link can be anywhere, and can be a linking topic - can be external or internal, so we can say 'in this placeholder we know it has to pull in .....' but we don’t know right now.

Cecil Lynch: What about version control?

Karen Smith: Yes a link can be internal (within this DITA document) or external to other documents

Paul Aneja: It would make the overall content much richer if we can refer to existing external or internal materials.

Karen Smith: Yes you can

Karen Smith: I plan to store the DITA files in SVN for version control

Cecil Lynch: Great

Jane Curry: We version the individual topic, so that as you update, you can replace the old with the new and not screw up everything else.

Grahame Grieve: of course

Ron G Parker: Yes, very good.

Karen Smith: yes

Karen - SVN is amenable to everyone.

Karen Smith: DITA is topic based rather than document based like Word

Karen Smith: great

Karen Smith: any more general questions

Ron G Parker: Yes, more questions coming.

Paul Aneja: as far as DITA goes, is it special software?

Jane Curry: Requires special software. Karen is using open source, and we may consider more elaborate tools.

Karen Smith: DITA is a specialized type of XML.

Ron Parker: Open source has a few restrictions.

Karen Smith: I would love to have more elaborate DITA tools, as I only have open source and freeware

Paul Aneja: Original question: ECCF is the first topic after the introduction, in the logical sense?

Ron G Parker: Karen, Jane has suggested for Phase 2 we might acquire more formal tools.

Karen Smith: Yes because ECCF is the most important part of SAEAF -- the specification stack.

Jane Curry: Yes. It introduces the specification stack. Then governance controls it

Karen Smith: Agree!!!!

John Koisch: Yes .... that was the proposal ... then IF and BF elaborate and drill down

Jane Curry: If you were to put in all in one book, that would be the way to introduce it. Now we have separate documents, with little recommendation on what to read first.

Paul Aneja: Do section overlap?

Jane Curry: The topic map will help control that.

Karen Smith: For the most part, no, but the WI info is pretty much the same in three of the four sections (don't know about Governance)

Karen Smith: Several of the graphics will be shared, though

Ron Parker: We will comment, but not edit the content.

Karen Smith: In DITA, you can reuse a topic if necessary in two different sections, such as the WI content or SS graphic

Jane Curry: She will be presenting the topic map. Once we get it straight, she will put it into the software.

Karen Smith: Yes

Karen Smith: Anyway, Jane, Cliff and I were talking about the audience of the SAEAF book. We came up with three major categories -- executives, architects, HL7 people. So the preface will explain about the audiences and why they would want to read about SAEAF

Karen Smith: And we decided to create an overall topic map that shows how SAEAF works that could go into the Introduction. I'll show you the map soon.

Jane Curry: When we go into the next phase we need to consider implementers as an audience.

Karen Smith: Yes, the implementers would be interested in the examples and implementation guide

John Koisch: Re-address implementers after phase 1

Karen Smith: Yes

Jane Curry: It is too technical for executives, and not deep enough for implementers.

John Koisch: The saeaf may not have that much to say to implementers. Topic specifiers may be encouraged to provide guides.

Karen Smith: Yes, we had a long discussion about the current executive overview in Deck 5. It's too technical for executives but more appropriate for architects. We were thking of having two different versions of the executive overview.

Jane Curry: We have to talk about it so people will achieve WI.

Karen Smith: To answer John's comment, we could provide links to guides for implementers in phase 2.

Jane Curry: for phase 2 we will actively consider implementers

Karen Smith: Yes that's the plan

Ron Parker: Provide P2 topic links to the documents.

Ron G Parker: Okay, Karen, this is great... I think we have consensus at this point, we can move on.

Karen Smith: Great, I will bring up the topic map.

Jane Curry: This is the high-level topic map (saeaf_comple_system_map.cmap)

Karen Smith: This is the overview of how SAEAF works. I would like to add it to the SAEAF Intro section just before the WI (value proposition) section. Any comments on this map, folks. You could email me the comments if you wish

John Koisch: Are we seeking approval of the mind map, or is it a reference ?

Jane Curry: Easily taking a picture of the topics.

John Koisch: It has relevance moving forward.

John Koisch: Is it published?

Karen Smith: no never been published -- created it yesterday

Jane Curry: On ARB list, she is using the features of the software to do the drilldown. You can provide comments on topics and links.

John Koisch: Things that need updating, or topics for ARB to discuss?

Karen Smith: does this map present an accurate high-level picture of SAEAF that doesn't get too technical?

Ron G Parker: Discussion going on... let's pause for a second

Jane Curry: She is working from existing material. When we started making maps, it made what was missing more obvious.

Jane Curry: Sometimes the sequence used does not make sense when you have the map.

Karen Smith: can you explain?

Ron Parker: Karen wants to know if this high level is ok?

Karen Smith: yes

Jane Curry: if the wording is not correct, or does not link well, we should comment. Once you look at threads it will remind you of something. We need topic maps within 2 weeks.

John Koisch: Information framework is missing

Karen Smith: yes, please send me your comments

Jane Curry: we need placeholders for things to be elaborated in Phase 2.

Cecil Lynch: Can you post the actual cmap file? One thing you can do with cmap is generate the RDF file that creates classes and properties that you can then use directly in an ontology.

Ron Parker: we need to help karen with the middle - the start and end are good.

Karen Smith: like "Information Framework (coming soon)

Karen Smith: great, Ron.

Karen Smith: These topic maps help me organize the tremendous amount of material

Jane Curry: Listserv dialog over the next two weeks.

Ron Parker: For next week we need synopsis to that point.

Karen Smith: I am on the listserv, so I can see the email

Jane Curry: Put comments in the e-mail on the listserv.

Karen Smith: yes

Ron Parker: Karen will monitor the listserv.

Karen Smith: will do

Paul Aneja: Links?

Cecil Lynch: Please also post the cmap file, not just pdf

Jane Curry: On the listserv will be published.

John Koisch: Less concerned about elaboration than the high level.

Karen Smith: Cliff will help with that

Jane Curry: If we cannot agree on the topics, we will never agree.

John Koisch: Difficult to express.

Ron Parker: Mind-map is contextual frame of reference - leveling and the logic path. We need to try to get that right. If it is coherent we have accomplished a great deal.

Jane Curry: It has changed in the last three days.

Karen Smith: yes, and I have been madly making outlines of all of the sections

Jane Curry: this i the equivalent map of the Introduction presentation (saeaf_intro.cmap). Drawn from the presentation, shows the overlaps and emphasis differences.

Jane Curry: we cant create a coherent book until we have a coherent topic map.

Ron Parker: We will get all to comment on?

Jane Curry: Yes?

Karen Smith: I now have the SAEAF Intro conceptual map up on the screen. I basically followed the organization of the SAEAF book with some exceptions -- moved the Jump Start stuff to SAEAF history, moved the easier concepts earlier in the document and the more difficult documents later. But a lot of the org is similar to the Intro Word doc.

Paul Aneja: Clickable?

Karen Smith: yes, we need a coherent topic map to create the outlines. Yes the cmap format is clickable

Jane Curry: we will make it available in the tool as well as the PDF.

Jane Curry: Did not want to make the tool a pre-requisite.

Karen Smith: ok

Karen Smith: Cliff will be setting up a cmap server for Jane and me to use.

Karen Smith: The cmap files are clickable as you saw

Cecil Lynch: http://cmap.ihmc.us/conceptmap.html

Ron Parker: Will it create a problem if we all wak at it?

Karen Smith: Cliff, can you answer that question?

Jane Curry: people use Skype chat or e-mail to ask, not edit the document.

Karen Smith: If you tweak the maps, save it under a different filename

John Koisch: I want to draw a topic map, and not change Karen’s stuff.

Karen Smith: sure you can draw your own topic map and send it to me

John Koisch: if we do it in paragraphs it is too wordy.

Ron Parker: use Skype or Arb list to comment. On the next call we will do an assessment. Jane, are there any other artifacts Karen needs to show us today?

Karen Smith: yes three more maps

Karen Smith: tada here is the governance map

Can we transition to the next agenda item? The project proposal by Lloyd.

Ron G Parker: Very nice Karen.

Jane Curry: We will follow the same process for the other maps - we're going to go on now

Karen Smith: okay you may take control now

Ron G Parker: We will provide feedback. John (and others) may build some topic maps

Cliff Thompson: very well done, Karen

Vocab WG email re: balloting the MIF

John Koisch: Lloyd is proposing we ballot the MIF. Jobst is pushing back - he is involved in a joint project OMG/HL7 to use UML. They are appearing at the NCI. He has strong ideas. What concerns me about his e-mail is it will open up other topics.

Ron G Parker: Will do. Thanks Karen.

Ron G Parker: Very nice work.

Ron G Parker: We are moving on to agenda item 2

John Koisch: The ArB has to say if the MIF is balloted, where it fits in the saeaf, and the thornier issue is the implications of the MIF ballot impacts the relationship with OMG.

Cliff Thompson: Tony, I can send you the particulars on CmapTools

John Koisch: The complete downside would be HL7, OMG disengaging. The relationship is not as smooth as it good be. It is ArB job to step in.

Ron Parker: Jane, Is that the HSSP proposal?

John Koisch: Just for HTTF, there are groups wanting to use SOAML, MIF, a group meeting in December around the BF - multiple streams.

Ron Parker: Concerns me: The proposal to ballot MIF is to lock it down before the 'disruptive' stuff goes to far.

Jane Curry: Case of responding to what is HL7's meta-model and why. He is providing rationale for the data elements, and there support of the methodology, and why.

Jane Curry: It is a case of exposing to the community the rules embedded in the tools - the methodology is less know today, than when we started. The HL7 methodology for producing contents.

John Koisch: Jobst is including the OMG's way of doing things - most of the reasons are that OMG thinks a certain way, not invalid, but not the HL7 way either. We need to have to dialog so we don’t become OMG.HL7. Arb must bring some reason to the arguments.

Grahame Grieve: There is no fear that balloting the MIF will lock it down - it is for internal review, should not affect the OMG relationship. I query the MIF balloting the set of schemas.

Grahame Grieve: Few people will be able to understand we are balloting an artifact to generate artifacts. The ArB should comment on the concrete vs abstract. It is about the meta-model, not the xml format.

John Koisch: This is another thing to bring to the ArB a parallel track for the BF - approved as an architectural artifact.

Cliff Thompson: Karen/Jane, i could do a one or two page doc on using the cmap tool for the group, if you want me to.

Karen Smith: that would be great, Cliff

Grahame Grieve: we should not interfere with the ballot, but should announce that it is in the wrong place.

Jane Curry: fine

Grahame Grieve: we should accept that the MIF ballot is pragmatic, going ahead.

Karen Smith: Is this MIF a file exchange format?

Cliff Thompson: okay, by Monday morning

Cecil Lynch: I favor that. It will allow it to make recommendation with the ballot process- separate the mif from the meta-model.

Grahame Grieve: Few can understand the MIF - it is a dreadful way of doing what we are trying to do.

Ron Parker: If we were to express a position, it would be guidance for balloters, expanding there awareness

Ron Parker: what is our vehicle as the Arb to make assertion - what and how?

John Koisch: The MIF is trying to do related to core-principles - what SAEAF calls information framework.

Grahame Grieve: in the end it will be superseded.

John Koisch: Since it is an informative ballot, open to looseness and interpretation, request it include links to the SAEAF book.

John Koisch: Plus we don’t have to do the work.

Grahame Grieve: Will not like doing that work

John Koisch: Architect is an imposition, and we are sorry.

Jane Curry: Not very

Grahame Grieve: Part is setting external parties to rest. Should we create a BLOG for ron to post, have RON post.

Karen Smith: Eventually I would need a list of external links (PDFs or web sites) that could go into the SAEAF book such as in a reference topic.

Jane Curry: Blog?

Grahame Grieve: yes.

Jane Curry: I don’t have problem, have no experience, it would be an enhancement. Is this a commitment to do it. No worse that stale-dated blog.

Karen Smith: and other documents could link to the SAEAF book when it is available

Ron Parker: ArB should have various communication.

Grahame Grieve: Use BLOG for informal 'press release'

Ron Parker: Likely with the activity, we might hit that 3-4 time between meetings, would be appropriate.

Jane Curry: work with marketing and communication.

Ron Parker: we don’t have communication plan.

Grahame Grieve: work item for this week: draft a communications strategy on the wiki, consult, get back to this next week ?

Ron Parker: Jane what are our precedence’s? does HL7 have a way to do this, without us spending cycles?

John Koisch: with the addition that we engage the MIF group to tie to Core Principles

John Koisch: (to Grahame ... FA)

Ron Parker: SAEAF seems to ratcheting up the Hype curve, without comment. I like the informal press release without it all thought out.

Jane Curry: we don’t have a communications office - we have people who produce press releases.

Ron Parker: Action item: Along the lines of using a blog for this purpose, I will talk to charlie and HL7.

Grahame Grieve: I will do the technical work for the Blog; draft a communications strategy on the wiki, consult, get back to this next week ?

Grahame Grieve: I will produce WIKI and send it out.

Jane Curry: run by John Quinn and TSC

Grahame Grieve: HQ and TSC.

Clarification on the new project scope statement for HSSP process in agreement with OMG and impact to the ArB.

Ron Parker: Jane, raise the HSSP thing?

Jane Curry: I wanted to know what the

John Koisch: Informally that project was undertaken by Galen because of inadequacies of the HSSP process in the work. To make sure all specifications are SAEAF compliant.

Jane Curry: Do we have a liaison to this project, so we can get feedback? t discuss issues,so we dont yell at the end of the project?

John Koisch: Galen is the liaison to HSSP.

John Koisch: Jane mentioned Governance ... Tony ... write that down

Jane Curry: It caught my eye because of what I saw in the governance framework, can we comment as it is done, or at the end.

Charlie Mead said "Jane Jane the Governance Gal"

Jane Curry: so we can get the lessons learned into the governance framework.

 Ron will invite Galen to the next call to that end.

Other business and planning for next meeting

Ron Parker: for next agenda: recap on DITA map: Galen for conversation: ECCF comment to the conformance group. Tony will link in the minutes.

Jane Curry: Outstanding question from the MAP compatibility is a good concept, but we have not defined it.

Karen Smith: also haven't defined LOCALIZATION

Ron Parker: Thanks all. Jane - pass to Karen that you narrating while she did the text worked well.

Karen Smith: thanks Tony and Jane for all your help during my presentation

Adjournment

Ron Parker: Meeting adjourned at 7:03pm U.S. EDT.